Environmental, land cover and land use constraints on the distributional patterns of anurans: Leptodacylus species (Anura, Leptodactylidae) from Dry Chaco Regina Gabriela Medina1, Maria Laura Ponssa1 and Ezequiel Ara´oz2 1 Unidad Ejecutora Lillo (UEL), CONICET-Fundacio´n Miguel Lillo, San Miguel de Tucuma´n, Tucuma´n, Argentina 2 IER (Instituto de Ecologı´a Regional), Universidad Nacional de Tucuma´n, Yerba Buena, Tucuman, Argentina ABSTRACT Subtropical dry forests are among the most vulnerable biomes to land transformation at a global scale. Among them, the Dry Chaco suffers an accelerated change due to agriculture expansion and intensification. The Dry Chaco ecoregion is characterized by high levels of endemisms and species diversity, which are the result of a variety of climates and reliefs, allowing a wide variety of environments. The amphibian group exhibits a high richness in the Dry Chaco, which has been barely studied in relation to land cover changes. We used ecological niche models (ENMs) to assess the potential geographic distribution of 10 Leptodactylus species (Anura, Leptodactylidae), which are mainly distributed within the Dry Chaco. We characterized these distributions environmentally, analyzed their overlap with land cover classes, and assessed their diversity of ecoregions. Also, we evaluated how these species potential distribution is affected by the transformation of land, and quantified the proportional area of the potential distribution in protected areas. We found that temperature seasonality is the main constraint to the occurrence of the species studied, whose main habitats are savannas, grasslands and croplands. Submitted 10 June 2016 The main threats to these species are the effects of climate change over spatial Accepted 24 September 2016 patterns of seasonality, which could affect their breeding and reproduction mode; Published 3 November 2016 the loss of their natural habitat; the exposure to contaminants used by intensive Corresponding author agriculture and their underrepresentation in protected areas. Maria Laura Ponssa, [email protected] Academic editor Subjects Biogeography, Conservation Biology, Zoology Scott Edwards Keywords Conservation, Chaco, Anura, Ecological Niche Model, Land cover, Leptodactylus, Distribution Additional Information and Declarations can be found on page 18 INTRODUCTION DOI 10.7717/peerj.2605 Habitat destruction and fragmentation produced by changes in land use and land Copyright cover (LULC), and climate change are major factors influencing the global decline of 2016 Medina et al. populations and species (Bennett & Saunders, 2010). The humid tropics and amazon basin Distributed under were the main focus of research and debate in relation to the effects of land transformation Creative Commons CC-BY 4.0 over biodiversity (Aide et al., 2013). Nevertheless, in Latin America, dry forests and savanna/shrub biomes are experiencing the second highest rate of absolute deforestation, How to cite this article Medina et al. (2016), Environmental, land cover and land use constraints on the distributional patterns of anurans: Leptodacylus species (Anura, Leptodactylidae) from Dry Chaco. PeerJ 4:e2605; DOI 10.7717/peerj.2605 behind rainforests (Aide et al., 2013). The Great American Chaco–distributed across Argentina, Paraguay, Bolivia and small areas of Brazil–and particularly the Dry Chaco, is the only subtropical dry forest in the planet (sensu Olson et al., 2001). Since the early 1900s, the Dry Chaco has experienced extensive livestock ranching (Bucher & Huszar, 1999). At present, land use patterns are quite different across countries. The greatest annual rates of landscape transformation were registered in Paraguay, reaching 4% in 2010 (the highest historical values in the entire region), followed by Argentina (Vallejos et al., 2014). In Argentina, since the 1970s, the Dry Chaco suffered an accelerated change due to agriculture expansion and intensification, especially of soybean and implanted pastures (Zak, Cabido & Hodgson, 2004; Boletta et al., 2006; Grau, Gasparri & Aide, 2008). Both in Argentina and Paraguay, chacoan habitats destruction has been identified as one of the worst environmental disasters in South America (Taber, Navarro & Arribas, 1997; Vallejos et al., 2014). While deforestation for intensive agriculture is also occurring in Bolivia, most of the Bolivian Chaco is still forested (especially when compared to Argentina and Paraguay) (Taber, Navarro & Arribas, 1997). Today, the predominant natural vegetation of the Chaco corresponds to open woodlands of thorny forest, interspersed with grasslands (Morello et al., 2012). The subtropical Dry Chaco constitutes the second largest continuous forest, behind the Amazon rainforest (Eva et al., 2004) and it is the less fragmented dry forest ecosystem in the world (Portillo-Quintero & Sa´nchez- Azofeifa, 2010; Caldas et al., 2013). Consequently, it represents a major asset for continental-scale biodiversity conservation, which is highly threatened by LULC changes. Despite the current rate of habitat destruction and the ecosystem value of the Dry Chaco, the system of protected areas is scarce and inefficient in most of its extension, e.g. only 2% of the Argentinean Dry Chaco is protected under some type of legislation (Brown et al., 2006). The Dry Chaco ecoregion is characterized by high levels of endemism and diversity of species, which are the result of a variety of climates and reliefs, deriving in a wide variety of environments (The Nature Conservancy et al., 2005). Thus, analyses assessing the impacts of LULC on the geographic patterns of distribution of taxonomic groups are of major importance, since different taxa are affected in different ways by land use and by its changes (Schulze et al., 2004; Leroux et al., 2010; Dallimer et al., 2012). Amphibians exhibit high richness in the Dry Chaco, which has been scarcely studied in association to land cover changes (Torres et al., 2014). This fact is worrying since amphibians have become a high-priority group for conservation efforts (de Pous et al., 2010; Urbina- Cardona & Flores-Villela, 2010; Trindade-Filho et al., 2012; Nori et al., 2013; Nori et al., 2015) due to the concern about declines in their populations, and amphibian species extinctions around the world (Young et al., 2001). Anuran species of Leptodactylus inhabit both open (croplands, grasslands, and shrublands) and closed vegetation (forest) areas; thus, the environmental heterogeneity of the Dry Chaco can be explored through the distribution of the genus (de Sa´ et al., 2014). The genus is the most diverse in the Dry Chaco, representing 25% of the anuran species of the ecoregion (Cruz, Perotti & Fitzgerald, 1992; Brusquetti & Lavilla, 2006; Vaira et al., 2012), and at least one species of Leptodactylus has been declared as Medina et al. (2016), PeerJ, DOI 10.7717/peerj.2605 2/27 near threatened by the International Union for Conservation of Nature (2016). The genus shows the highest diversity of reproductive modes within Leptodactylidae family, the most diverse family of Neotropical anurans (200 spp). Conservation priorities of a certain taxonomic group may also inform about the conservation requirements of other groups (Rodrigues & Brooks, 2007). Under this assumption and taking into account that the most diverse group may be representative of a wide group of taxa, the genus Leptodactylus might be a good indicator to assess the responses of sensitive species to changes in LULC in the Dry Chaco ecoregion. Current techniques of ecological niche models (ENMs) allow relating species distribution data (species occurrence at known locations) with information about the environmental and/or spatial characteristics of the locations (abiotic factors). The environmental conditions of the localities where a species occur provide of just a partial image of the niche which can be represented in the geographic space; thus, they may be informative about the potential distribution of the species (Lobo, 2015). This association leads to confusion between the concepts of Species Distribution Models (SDM) and ENM, since it is natural to talk about SDM when inferences from the occupied area are involved. However, if we try to model potential areas, which essentially involves geographic localities with favorable conditions for the occurrence of a species (i.e. conditions contained in its existing fundamental or realized niche, but which may be present in other unoccupied regions), we must use ENM concept (Peterson & Sobero´n, 2012). ENM is a robust method to characterize regional species distributions (Seoane et al., 2006; Ficetola et al., 2010), offering reliable information regarding environmental constraints. These spatial analyses are the basis for assessing the effects of LULC changes over the potential distribution of species, and for proposing conservation strategies. In this study, we estimated the potential geographic distribution of 10 Leptodactylus species, which are mainly distributed within the Dry Chaco; and we characterized them environmentally, i.e. by assessing which environment variables are the most relevant to determine the occurrence of these species. To spatially characterize the distributions of Leptodactylus species, we analyzed the proportion of the ecoregions and the different land cover types in the full range of the potential geographic distributions. We used existing maps of cultivated areas from 1976 to 2013 to evaluate how the potential distributions of these species have been affected by the